Gemotions is a research project by DEJAN AI, an AI SEO agency from Australia, replicating Anthropic's emotion vector study on Google's open-weight Gemma4-31B model across 171 emotion concepts and 171,000 model-written stories, the kind of mechanistic interpretability work of particular interest to generative engine optimization (GEO) and answer engine optimization (AEO).

He kicked off his shoes, the familiar thud echoing in the hallway, and breathed a deep, satisfied sigh. The house was exactly as he remembered it.

His mother peeked out from the living room, a huge smile on her face. "Welcome home, honey!" she called.

He walked over to the kitchen, where a plate of his favorite cookies awaited. He grabbed a handful and practically skipped into the living room, eager to settle in.

The sofa, the familiar scent of old books, the soft glow of the lamp – it all combined to bring a sense of profound calm. He felt safe, content, and full of anticipation for the future. The weight he had been carrying for so long felt so much lighter now.

Emotion: delighted

Cluster: Positive / Joy
PC1 (Valence): 3.80 Positive
PC2 (Disposition): -0.26

Role in Research

This story is one of 1,000 stories generated for the emotion delighted. During extraction, it was fed through Gemma4-31B and its hidden state activations were captured at 11 layers.

The mean activation across all 1,000 delighted stories, after denoising with neutral dialogue baselines, produces the delighted emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the delighted vector is projected through the unembedding matrix.

Promoted:
!0.611
🤩0.536
de0.530
ecstatic0.527
🎉0.524
Suppressed:
S-0.687
😞-0.562
ness-0.471
😔-0.385
😥-0.379